Minimization of energy consumption is critical to developing green, sustainable technologies for cognitive radio terminals that can connect to networks that operate on different frequency bands with a variety of air interfaces. The intellectual merit of this project is a unified and coherent consideration of RF components, communication system algorithms, baseband computation platforms, and design tools, to greatly increase spectrum sharing efficiency. Dataflow methodologies are a promising candidate for the modeling, analysis and verification of cognitive radio systems. As dataflow models are abstract and platform independent, the same model can be used to generate implementations for very different devices from low-power sensor nodes to high-end mobile terminals. The key novelty is in the development of systematic methods for design, implementation, and integration of configurable RF chains, and in the development of dataflow methods for formal analysis and optimization of these new capabilities. The expected results are: (1) Energy consumption models and a design framework for computation, control and configuration of future radio devices, leveraging the investigators' existing experimental testbeds, (2) Configurable radio architectures for wide-scale cognitive access of noncontiguous RF spectrum, and (3) Design methodologies for flexible, energy-efficient cognitive wireless networks. The broader impact includes international collaboration through the WiFiUS program creating a holistic design for configurable frequency agile terminals. A novel interdisciplinary approach is enabled by the unique international team, which builds upon collaborations between experts at the Tampere University of Technology and University of Oulu in Finland, and Rice University and the University of Maryland in the US.
